Hi Tim,

  There is no capability to get revised pLDDT scores in ChimeraX.  Those predicted local distance difference test scores are computed by AlphaFold 3 based as part of predicting the structure.  If you wanted to remove disordered N and C-terminal tails of proteins in your predicted complex you could run a new prediction where those parts of the input sequences are removed.

  For judging interface prediction confidence many people use the ipTM (interface predicted TM) scores computed by AlphaFold 3.  For instance in output file

fold_2026_07_15_12_10_summary_confidences_0.json

for a 4 chain prediction I see the following 4 by 4 matrix of ipTM scores in a text editor

 "chain_pair_iptm": [
  [
   0.81,
   0.84,
   0.64,
   0.72
  ],
  [
   0.84,
   0.41,
   0.03,
   0.02
  ],
  [
   0.64,
   0.03,
   0.71,
   0.05
  ],
  [
   0.72,
   0.02,
   0.05,
   0.04
  ]
 ],

The ipTM scores are computed from the AlphaFold 3 predicted aligned error (PAE) scores.  The PAE scores give confidence for every pair of residues.  I prefer to look directly at the PAE scores in ChimeraX.  Here is an explanation of how to show that

https://www.rbvi.ucsf.edu/chimerax/data/alphafold-contacts-jun2022/afcontacts.html

and here is a video I made showing how to look at the interface confidence in ChimeraX

https://www.youtube.com/watch?v=TMcjEecFHaI

  Tom

Image shows AlphaFold confidence for interface between green and pink proteins.  Blue lines between residues indicate high confidence and yellow indicate low confidence.  This is a more fine-grain view of confidence than chain pair ipTM scores.

pae_contacts.png



On Aug 17, 2026, at 5:08 PM, Timothy Springer via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:

The PAE viewer in Chimerax is wonderful. I can delete residues with low confidence pLLDT in the AF3 model (B factor <55) and the PAE viewer still works (only problem is deleting whole chains). 
Is there a way to delete the same residues and get revised pLLDT scores? I want them per interface in a multichain complex. 

- Tim
Timothy A. Springer, Ph.D.
Latham Family Professor, Harvard Medical School and Boston Children's Hospital
Founder, Institute for Protein Innovation
_______________________________________________
ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu
To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu
Archives: https://mail.cgl.ucsf.edu/mailman/archives/list/chimerax-users@cgl.ucsf.edu/